Re: Bug#2345: Inclusion of shadow breaks up some packages in the current (0.93R6) packaging system
> David> But if I want to contribute a shadow-3.3.2-1.deb-Package, it's
> David> necessary that the _distribution_ has the packages shellutils and
> David> miscutils broken up into an additional package `login' as well, so
> David> that I'm able to _replace_ the old login, passwd, su stuff with the
> David> new shadow programs.
> Right, but did you see the discussions on debian-devel on this? I am not sure
> that there is an established consensus to go for shadow-3.3.2 *now*. Maybe
I'm sorry, but which discussion? Also I cannot see the problem. It shouldn't
be too much work to create a login_utils package. No one is forced to use
the shadow package, but at least we give the users the possibility to do so.
So, where's the problem?
> David> It would be nice to have shadow passwords as an option, this would
> David> be an big plus to choose Debian over slackware/redhat or whatever
> David> (besides the packaging system... :)
> I do personally agree, I used to use shadow passwords. But I *really* what
> the transition to happen in an organised manner, and that involves a little
> more I am afraid (ie recompiling/linking of netstd, wu-ftpd, xlockmore, ...)
Yes, there are more packages involved, but, hey, we're talking about the
development tree. I cannot see a reason why we cannot have recompiled
versions of these programs in the shadow package, too. It's better though to
create additional packages for these binaries. I do use shadow and compiling
these binaries for shadow support is not much work to do. Also, there's a
way to get the shadow password via the standard libc. Just look into
Miquel's sulogin program. It works fine for me (with shadow passwords)
although I didn't have to do any changes.
BTW there's a another position that needs changing: The config files for
Anyway, I can only agree to David, that an inclusion of teh shadow suite
would be a very big plus for Debian over all the other distributions. And I
bet come time others will include it, too. Why not being the first?
Lehrstuhl fuer angewandte Mathematik insb. Informatik
RWTH-Aachen, D-52056 Aachen, Germany